Most of our mechanistic understanding of how human cancer cells migrate and invade has been obtained by observing cell behavior in an artificial 2 D environment. Although progress has been made using this approach, important new evidence indicates that cell migration in 2 D systems does not completely recapitulate events associated with locomotion in a more physiological environment using reconstituted 3 D matrices and tissue explants. Work by others and novel evidence provided in this research proposal demonstrate invasive cells can utilize either a mesenchymal type of cell invasion that involves formation of an elongated invadapodia and a spindle shaped morphology or a primitive amoeboid movement that involves membrane blebbing though small holes in the extracellular matrix. These breakthrough findings prompted the hypothesis that cells are armed with different invasive programs that allow them to traverse complex tissues and colonize foreign sites in the body. Most importantly though these findings indicate that therapeutic prevention of this process in patients will require a multifaceted approach that targets both modes of cell invasion. It is crucial then that we identify invasive mechanisms utilized by disseminating tumor cells in vivo so that the appropriate therapeutic agent(s) can be designed to completely eradicate the spread of cancer in patients. However, tumor cell invasion is a complex and dynamic process that involves the intricate interplay between the tumor cells and the remodeling vasculature and stroma. Understanding this process in vivo has been difficult because it has not been possible to visualize this process in high resolution in live animals. To address this problem, we have developed a novel model of cancer progression that utilizes human cancer cells growing in optical clear zebrafish genetically engineered to express green fluorescent protein in all blood vessels. Using this model and dual color high resolution confocal microscopy, we discovered that the metastatic gene RhoC induces a rapid cell invasion process that facilitates cell intravasation through vascular openings induced by VEGF secretion. In contrast, mesenchymal cell invasion involves formation of elongated invadopodia and membrane integration into the vascular wall, but not cell intravasation. Our goal in the proposed work is to understand the signaling mechanism that control amoeboid and mesenchymal invasion as cells intravasate and how the vascular pores form in response to VEGF secretion. Based on our preliminary findings and the work of others, we hypothesize that the metastatic gene RhoC mediates amoeboid invasion through Rho kinase activity (ROCK) and myosin II-mediated contractility. We also hypothesize that PI3K harboring activating mutations found in human cancers induces mesenchymal cell invasion through activation of the FAK-Src-CAS-Crk-Rac signaling module, which facilitates actin-mediated invadopodial protrusion. We hypothesize that the vascular pores form through disruption of cell-cell junctions, which is regulated by src phosphorylation of VE-cadherin. Therefore, our overall goal is to examine in detail how RhoC and mutated PI3K signaling pathways regulate cancer cell invasion and intravasation and the molecular signaling mechanisms that control vascular pore formation.